Hello,
Can anyone explain why the following message appears for the function
model.tables, where se=T? In the V&R MASS text, p.285 the se=T option works
for a split plot example that seems similar to my operation. But the
model.tables documentation, in the Arguments section for "se", states
"should standard errors be computed?".
"Warning: Warning in model.tables.aovlist(Magic.aov, type =
"means", se=T):
SE's for type 'means' are not implemented yet"
I am trying to get the Standard Errors (SE's) for a split plot using the aov
function. I'm not sure, it may be a split-split plot design. There are two
crossed factors: 1)Formula 2)SwatchType.
* Formula is applied to the whole plots. WP = Load.
* Each Whole plot is further divided into 3 split plots. SP = BackrNbr, 3
BackrNbrs within a single Load.
* Each split plot is the block for a completely randomized design using factor =
SwatchType. All 6 SwatchTypes are applied within each BackrNbr
Magic.aov<-aov(SoilRemoval~Formula*SwatchType+Error(Load/BackrNbr),data=Magic.dat)
model.tables(Magic.aov, type="means",se=T) ### the table of means are
printed, no SE's
> str(Magic.dat)
'data.frame': 162 obs. of 7 variables:
$ BackrNbr : Factor w/ 27 levels
"1","2","3","4",..: 1 1 1 1 1 1 2 2 2 2
...
$ Load : Factor w/ 9 levels
"1","2","3","4",..: 1 1 1 1 1 1 1 1 1 1
...
$ Formula : Factor w/ 3 levels "CONTROL","EXP1",..: 2 2
2 2 2 2 2 2 2 2 ...
$ SwatchType : Factor w/ 6 levels "DMO","Dust.Seb",..: 2 6
1 5 4 3 2 6 1 5 ...
$ L.Initial : num 71.2 73.9 69.3 58.6 40.6 ...
$ L.Final : num 89.8 75.6 70.5 69.6 58.9 ...
$ SoilRemoval: num 75.08 7.88 4.71 29.44 33.15 ...
Thank you for your consideration,
Paul
Paul Prew | Statistician
651-795-5942?? |?? fax 651-204-7504
Ecolab Research Center | Mail Stop ESC-F4412-A
655 Lone Oak Drive | Eagan, MN 55121-1560
CONFIDENTIALITY NOTICE: =\ \ This e-mail communication a...{{dropped:12}}